- Email: [email protected]
El Niño and variations in the prevalence of Plasmodium vivax and P. falciparum in Vanuatu
Transactions of the Royal Society of Tropical Medicine and Hygiene (2009) 103, 1285—1287
available at www.sciencedirect.com
journal homepage: http://www.elsevier.com/locate/trstmh
SHORT COMMUNICATION
El Ni˜ no and variations in the prevalence of Plasmodium vivax and P. falciparum in Vanuatu M. Gilbert a, R. Brindle b,∗ a b
Imperial College London, UK Portsmouth Hospitals, Portsmouth, UK
Received 1 October 2008; received in revised form 29 October 2008; accepted 29 October 2008 Available online 13 December 2008
KEYWORDS Malaria; Plasmodium vivax; Plasmodium falciparum; Prevalence; El Ni˜ no; Vanuatu
Summary Malaria, both Plasmodium falciparum and P. vivax, is a major cause of morbidity in Vanuatu. As P. vivax is more prevalent in seasonal climates and P. falciparum in areas of more consistent rainfall, it is postulated that there will be a correlation between the ratio of vivax:falciparum and the El Ni˜ no Southern Oscillation (ENSO), which affects sea surface temperatures and rainfall. With changes in global climate, the frequency, duration and strength of the ENSO are expected to alter, influencing the pattern of malaria. The data showed no obvious correlation between ENSO and either cases of malaria or the vivax:falciparum ratio. © 2008 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.
1. Introduction Malaria is a major cause of morbidity in Vanuatu and was reported as the major cause of death in children after the neonatal period in 19881 and a significant cause of morbidity.2 Both Plasmodium falciparum and P. vivax are prevalent in Vanuatu, with transmission being more intense in the north. The El Ni˜ no Southern Oscillation (ENSO) is a global atmospheric oscillation in temperature and pressure between the Pacific and Indo-Australian areas and is closely related to the periodic warming (El Ni˜ no) and cooling (La Ni˜ na) of the eastern equatorial pacific. This has consequences on
∗
Corresponding author. Present address: Queen Alexandra Hospital, Portsmouth, PO6 3LY, UK. Tel.: +44 2392 286 000. E-mail address: [email protected] (R. Brindle).
public health, with good evidence that it affects malaria rates.3 Plasmodium vivax is the prevalent strain of malaria in seasonal climates (marked dry and wet seasons) and P. falciparum in areas of more consistent rainfall. It is postulated that the variations in total numbers of patients with parasitaemia and the ratio of vivax to falciparum will be correlated with both monthly temperatures and rainfall, which can be related to ENSO. With changes in global climate the frequency, duration and strength of ENSO are expected to be altered. These alterations are likely to influence the pattern of malaria.4
2. Materials and methods Monthly figures on malaria, rainfall and temperature were collected. The malaria data were collected retrospectively from hospital records on three islands: Efate (centre), Tanna (south) and Santo (north). Access to hospital-held data was
0035-9203/$ — see front matter © 2008 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.trstmh.2008.10.048
1286
M. Gilbert, R. Brindle
Figure 1 The ratio of Plasmodium vivax to P. falciparum (diamonds) cases plotted with the southern oscillation index (SOI) (squares).
granted by the malaria pathologist with permission from the Vanuatu Government. Malaria data were recorded as numbers of cases of malaria per month grouped according to the type of parasite present. This was done over an 11 year period from 1995 to 2005.
2.1. Meteorological data Monthly mean rainfall and air temperature recorded for the 11 year period between 1995 and 2005, inclusive, was provided by the Government station on Efate, which records local meteorological information, as well as collating data from other sites, including Tanna and Santo. Southern oscillation index data was taken from the Australian Bureau of Meteorology.5
3. Results 3.1. Malaria Annual and monthly totals were calculated for both P. falciparum and P. vivax. Mixed infections and presumptive P. malariae infections were ignored as they formed 0.22 and 0.20% of recorded infections, respectively. The ratios of P. vivax to P. falciparum were calculated for years and months.
3.2. Meteorology Rainfall and temperature were very closely associated, with the greatest rainfall occurring at times of highest temperature. Over the 11 year observation period there appears to have been an overall rise in temperature of about 0.5 ◦ C. Annual rainfall has varied, almost twofold over the 11 years. There was a clear positive correlation between the amount of rain each month and the number of cases of malaria. Over the observation period the number of recorded cases of malaria increased, particularly P. vivax, and was not clearly associated with rainfall.
There was no clear correlation between years with high rainfall and the proportion of cases of vivax compared to falciparum. For example, 1997 was a relatively dry year (rainfall 1489 mm) and there were 1.2 times more cases of falciparum than vivax, but in 2004, also dry (rainfall 1505 mm), there were 2.4 times as many cases of vivax as falciparum. There was no obvious correlation between local rainfall and the southern oscillation index (SOI), and there was no obvious correlation between the SOI and the numbers of patients presenting to the three hospitals with malaria (Figure 1).
4. Discussion Although there was a very close relationship between seasonal rainfall and numbers of cases of malaria, there was no clear relationship between the species of malarial parasite and rainfall. It might be expected that with increased rainfall transmission of malaria would be more efficient, and this would favour an increase in cases of falciparum over those of vivax. However, the data were dominated by cases of malaria from Efate and Santo, where P. falciparum is hyperendemic, with transmission occurring all year round. Tanna has very few cases of falciparum recorded, with episodic epidemics, which are not obviously related to rainfall and probably are associated with periodic importation of cases from other islands. This study has highlighted the complexity of malaria transmission and suggests that in an area of already high malaria endemicity small variations in ENSO are not obviously responsible for variations in the total number of malaria cases, nor the relative incidence of the different malaria species. These changes are more likely to be a reflection of the efficiency of local control measures, population movements, variations in the provision of microscopy services and other social factors. Further study with more controlled variables would be useful, as climate change appears inevitable, and this will undoubtedly affect the transmission and control of malaria.
El Ni˜ no and variations in the prevalence of Plasmodium vivax and P. falciparum in Vanuatu Authors’ note: The complete data are available from the authors.
1287
Ethical approval: Not required.
References Authors’ contributions: MG and RB designed the study protocol; MG carried out the data collection; RB and MG carried out analysis and interpretation of these data. MG and RB drafted the manuscript. Both authors read and approved the final manuscript. MG and RB are guarantors of the paper. Acknowledgements: The authors would like to acknowledge the assistance of R. Tigona from the Vanuatu Meteorology Department, who provided rainfall and sea temperature data, as well as M. Amos and G. Taleo of the Vanuatu Government for their advice on data collection. The authors would also like to acknowledge the assistance with data collection given by J. Bowen. Funding: None. Conflicts of interest: None declared.
1. Reeve PA. Malaria treatment in Vanuatu: new national treatment guidelines. PNG Med J 1994;37:181—8. 2. Khazei A, Jarvis-Selinger S, Ho KL. An assessment of telehealth needs and health-care priorities of Tanna Island: a remote, under-served and vulnerable population. J Telemed Telecare 2005;11:35—40. 3. Kovats RS, Bouma MJ, Hajat S, Worrall E, Haines A. El Ni˜ no and health. Lancet 2003;362:1481—9. 4. EOSDIS Data Sampler #3 ENSO effects: malaria. Maryland: NASA EOSDIS Science Operations Office. http://outreach. eos.nasa.gov/EOSDIS CD-03/docs/enso malaria.htm [accessed 28 October 2008]. 5. S.O.I. (Southern Oscillation Index) Archives. Melbourne, Victoria: Australian Government Bureau of Metereology. http://www.bom.gov.au/climate/current/soihtm1.shtml [accessed 27 November 2008].
available at www.sciencedirect.com
journal homepage: http://www.elsevier.com/locate/trstmh
SHORT COMMUNICATION
El Ni˜ no and variations in the prevalence of Plasmodium vivax and P. falciparum in Vanuatu M. Gilbert a, R. Brindle b,∗ a b
Imperial College London, UK Portsmouth Hospitals, Portsmouth, UK
Received 1 October 2008; received in revised form 29 October 2008; accepted 29 October 2008 Available online 13 December 2008
KEYWORDS Malaria; Plasmodium vivax; Plasmodium falciparum; Prevalence; El Ni˜ no; Vanuatu
Summary Malaria, both Plasmodium falciparum and P. vivax, is a major cause of morbidity in Vanuatu. As P. vivax is more prevalent in seasonal climates and P. falciparum in areas of more consistent rainfall, it is postulated that there will be a correlation between the ratio of vivax:falciparum and the El Ni˜ no Southern Oscillation (ENSO), which affects sea surface temperatures and rainfall. With changes in global climate, the frequency, duration and strength of the ENSO are expected to alter, influencing the pattern of malaria. The data showed no obvious correlation between ENSO and either cases of malaria or the vivax:falciparum ratio. © 2008 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.
1. Introduction Malaria is a major cause of morbidity in Vanuatu and was reported as the major cause of death in children after the neonatal period in 19881 and a significant cause of morbidity.2 Both Plasmodium falciparum and P. vivax are prevalent in Vanuatu, with transmission being more intense in the north. The El Ni˜ no Southern Oscillation (ENSO) is a global atmospheric oscillation in temperature and pressure between the Pacific and Indo-Australian areas and is closely related to the periodic warming (El Ni˜ no) and cooling (La Ni˜ na) of the eastern equatorial pacific. This has consequences on
∗
Corresponding author. Present address: Queen Alexandra Hospital, Portsmouth, PO6 3LY, UK. Tel.: +44 2392 286 000. E-mail address: [email protected] (R. Brindle).
public health, with good evidence that it affects malaria rates.3 Plasmodium vivax is the prevalent strain of malaria in seasonal climates (marked dry and wet seasons) and P. falciparum in areas of more consistent rainfall. It is postulated that the variations in total numbers of patients with parasitaemia and the ratio of vivax to falciparum will be correlated with both monthly temperatures and rainfall, which can be related to ENSO. With changes in global climate the frequency, duration and strength of ENSO are expected to be altered. These alterations are likely to influence the pattern of malaria.4
2. Materials and methods Monthly figures on malaria, rainfall and temperature were collected. The malaria data were collected retrospectively from hospital records on three islands: Efate (centre), Tanna (south) and Santo (north). Access to hospital-held data was
0035-9203/$ — see front matter © 2008 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.trstmh.2008.10.048
1286
M. Gilbert, R. Brindle
Figure 1 The ratio of Plasmodium vivax to P. falciparum (diamonds) cases plotted with the southern oscillation index (SOI) (squares).
granted by the malaria pathologist with permission from the Vanuatu Government. Malaria data were recorded as numbers of cases of malaria per month grouped according to the type of parasite present. This was done over an 11 year period from 1995 to 2005.
2.1. Meteorological data Monthly mean rainfall and air temperature recorded for the 11 year period between 1995 and 2005, inclusive, was provided by the Government station on Efate, which records local meteorological information, as well as collating data from other sites, including Tanna and Santo. Southern oscillation index data was taken from the Australian Bureau of Meteorology.5
3. Results 3.1. Malaria Annual and monthly totals were calculated for both P. falciparum and P. vivax. Mixed infections and presumptive P. malariae infections were ignored as they formed 0.22 and 0.20% of recorded infections, respectively. The ratios of P. vivax to P. falciparum were calculated for years and months.
3.2. Meteorology Rainfall and temperature were very closely associated, with the greatest rainfall occurring at times of highest temperature. Over the 11 year observation period there appears to have been an overall rise in temperature of about 0.5 ◦ C. Annual rainfall has varied, almost twofold over the 11 years. There was a clear positive correlation between the amount of rain each month and the number of cases of malaria. Over the observation period the number of recorded cases of malaria increased, particularly P. vivax, and was not clearly associated with rainfall.
There was no clear correlation between years with high rainfall and the proportion of cases of vivax compared to falciparum. For example, 1997 was a relatively dry year (rainfall 1489 mm) and there were 1.2 times more cases of falciparum than vivax, but in 2004, also dry (rainfall 1505 mm), there were 2.4 times as many cases of vivax as falciparum. There was no obvious correlation between local rainfall and the southern oscillation index (SOI), and there was no obvious correlation between the SOI and the numbers of patients presenting to the three hospitals with malaria (Figure 1).
4. Discussion Although there was a very close relationship between seasonal rainfall and numbers of cases of malaria, there was no clear relationship between the species of malarial parasite and rainfall. It might be expected that with increased rainfall transmission of malaria would be more efficient, and this would favour an increase in cases of falciparum over those of vivax. However, the data were dominated by cases of malaria from Efate and Santo, where P. falciparum is hyperendemic, with transmission occurring all year round. Tanna has very few cases of falciparum recorded, with episodic epidemics, which are not obviously related to rainfall and probably are associated with periodic importation of cases from other islands. This study has highlighted the complexity of malaria transmission and suggests that in an area of already high malaria endemicity small variations in ENSO are not obviously responsible for variations in the total number of malaria cases, nor the relative incidence of the different malaria species. These changes are more likely to be a reflection of the efficiency of local control measures, population movements, variations in the provision of microscopy services and other social factors. Further study with more controlled variables would be useful, as climate change appears inevitable, and this will undoubtedly affect the transmission and control of malaria.
El Ni˜ no and variations in the prevalence of Plasmodium vivax and P. falciparum in Vanuatu Authors’ note: The complete data are available from the authors.
1287
Ethical approval: Not required.
References Authors’ contributions: MG and RB designed the study protocol; MG carried out the data collection; RB and MG carried out analysis and interpretation of these data. MG and RB drafted the manuscript. Both authors read and approved the final manuscript. MG and RB are guarantors of the paper. Acknowledgements: The authors would like to acknowledge the assistance of R. Tigona from the Vanuatu Meteorology Department, who provided rainfall and sea temperature data, as well as M. Amos and G. Taleo of the Vanuatu Government for their advice on data collection. The authors would also like to acknowledge the assistance with data collection given by J. Bowen. Funding: None. Conflicts of interest: None declared.
1. Reeve PA. Malaria treatment in Vanuatu: new national treatment guidelines. PNG Med J 1994;37:181—8. 2. Khazei A, Jarvis-Selinger S, Ho KL. An assessment of telehealth needs and health-care priorities of Tanna Island: a remote, under-served and vulnerable population. J Telemed Telecare 2005;11:35—40. 3. Kovats RS, Bouma MJ, Hajat S, Worrall E, Haines A. El Ni˜ no and health. Lancet 2003;362:1481—9. 4. EOSDIS Data Sampler #3 ENSO effects: malaria. Maryland: NASA EOSDIS Science Operations Office. http://outreach. eos.nasa.gov/EOSDIS CD-03/docs/enso malaria.htm [accessed 28 October 2008]. 5. S.O.I. (Southern Oscillation Index) Archives. Melbourne, Victoria: Australian Government Bureau of Metereology. http://www.bom.gov.au/climate/current/soihtm1.shtml [accessed 27 November 2008].